Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →There is no single best pseudopotential for every Quantum ESPRESSO calculation. Choose a file that matches your exchange-correlation functional and required code features, then check its relativistic treatment and valence states. Use its suggested cutoffs as starting points and converge the properties you plan to report. The Quantum ESPRESSO pseudopotential page recommends curated collections such as SSSP; still test a candidate on simple systems before relying on it.
Start with the calculation, not the file
Write down the elements and chemical environments you will model, the target property, the exchange-correlation (XC) functional, and the Quantum ESPRESSO executable or package you intend to use. A file that works well for one material or observable may not be suitable for another: the choice balances transferability, accuracy, code compatibility, and computational cost. The Quantum ESPRESSO pseudopotentials FAQ advises choosing based on the transferability and efficiency required by the calculation.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Computational Physics | $37.05 | Buy on Amazon |
| 2 |
|
Physics for Scientists and Engineers With Modern Physics | $91.26 | Buy on Amazon |
| 3 |
|
Computational Physics | $19.73 | Buy on Amazon |
| 4 |
|
Advanced Physics Problems And Solutions - Volume 1: Classical Mechanics And Mathematical Methods... | $38.00 | Buy on Amazon |
| 5 |
|
A Computational Introduction to Quantum Physics | $34.43 | Buy on Amazon |
As an Amazon Associate I earn from qualifying purchases.
Consider the observable explicitly. Equilibrium structures, energy differences, forces, stress, phonons, and spin-orbit splittings can place different demands on a dataset. Converge the quantities that matter to your result rather than assuming a file is reliable because it produced a plausible energy or structure.
Check that the pseudopotential family supports your calculation
Quantum ESPRESSO uses UPF files for norm-conserving (NC), ultrasoft (USPP), and projector augmented-wave (PAW) datasets. These families are not interchangeable in every calculation path. The official pseudopotentials page lists examples of feature constraints: meta-GGA, Gamma-only phonons, and third-order energy derivatives are among features that work only with NC datasets, while Car-Parrinello (CP) does not yet support PAW.
#1 Best Overall
Check the current documentation for the specific feature, executable, and software version you plan to use before selecting a family. Do not infer compatibility just because a UPF file can be read by a different QE calculation.
Match the functional and inspect the UPF metadata
Prefer a pseudopotential generated for the XC functional used in the calculation. A filename abbreviation is not enough to establish a match: open the UPF and inspect its metadata. The UPF format documentation describes fields including the functional label (dft), valence charge (Zval), family flags (is_uspp and is_paw), spin-orbit information (has_so), nonlinear core correction (nlcc), and suggested ecutwfc and ecutrho.
Official PSLibrary tables illustrate how many variants can exist even for one element. The silicon table lists PBE and PBEsol options, PAW and USPP constructions, and scalar- and fully relativistic files. The beryllium table also contains choices that vary by functional, construction, and relativistic treatment. Treat the exact file—not just the element name—as the unit you evaluate.
Choose relativistic treatment and valence states for the physics
Scalar-relativistic or fully relativistic
If spin-orbit effects are part of the target, select fully relativistic data with spin-orbit information and confirm that your chosen QE calculation path supports its use. A scalar-relativistic file is not a substitute when the result depends on spin-orbit splitting. For calculations where those effects are not being modeled, the appropriate choice depends on the material, target accuracy, and available validated datasets. The UPF metadata and the PSLibrary silicon entries can help distinguish scalar and fully relativistic candidates.
Rank #3
Valence and semicore states
Check the valence charge and the states included in the dataset. Frozen-core choices can limit transferability when the modeled chemistry or electronic configuration differs from the configurations for which a dataset performs well. Semicore states can matter for some elements and properties, but they are not a universal requirement.
For example, Quantum ESPRESSO’s pseudopotential-generation guide discusses a titanium dataset with 3d, 4s, and 4p states whose transferability across different 3d configurations was limited; it considers including 3s and 3p semicore states. That example motivates checking Ti candidates for your chemistry—it does not mean every Ti calculation must use that valence configuration.
Compare candidates on the factors that affect your result
| What to compare | What to inspect | Why it matters |
|---|---|---|
| Feature support | NC, USPP, or PAW family; target executable and property | Some QE features have family-specific constraints. Check the official compatibility notes. |
| Functional | UPF dft label against the calculation’s XC functional |
A mismatch can make the setup inconsistent; see the QE FAQ and UPF metadata documentation. |
| Relativity | Scalar or fully relativistic treatment and spin-orbit data | The file must represent the effects your calculation aims to model. |
| Valence and transferability | Valence charge, included states, semicore choices, and tested configurations | Frozen-core assumptions may affect performance across chemical environments. |
| Numerical cost | File-specific suggested cutoffs and converged cutoffs | Cutoffs affect plane-wave cost; density cutoffs also depend on the family. |
| Evidence and provenance | Source library, validation information, file version, and authorship | These details help establish suitability, reproducibility, and proper credit. |
SSSP is a sensible starting point because QE recommends it as a curated collection of verified pseudopotentials. QE also documents PSLibrary and other ready-to-use tables on its pseudopotentials page. A library’s validation does not guarantee that every entry suits every material, property, or feature requirement; compare candidates against your use case.
Set and converge plane-wave cutoffs
Begin with the ecutwfc and ecutrho suggested in the specific UPF file, then test higher values. These suggestions belong to that file; they are not universal settings for the element or family. The QE pw.x input reference gives practical guidance: USPP often benefits from an ecutrho about 8–12 times ecutwfc; NC generally uses the default ratio; and PAW density cutoffs depend on augmentation charge, so testing is required.
Best Value
- Set an initial pair. Read both suggested cutoffs from the candidate UPF and use them as the start of a convergence test.
- Increase the cutoffs. Recalculate at successively higher settings while holding other inputs fixed. Track the property you will report, such as energy differences or forces; include stress if it affects your conclusions.
- Choose a practical converged setting. Set an accuracy tolerance appropriate to the scientific question, then use cutoffs at which the target observable is stable within that tolerance.
- Converge k-point sampling separately. A cutoff test does not establish k-point convergence. Test sampling independently so you can identify which setting is limiting the result.
A historical QE carbon test file shows ecutwfc values of 24, 26, 28, 30, and 32 Ry and ecutrho values of 160, 200, and 240 Ry for stated graphite and diamond calculations. Those are examples for the specified file and systems, not recommended defaults for current carbon calculations or other elements. See the carbon convergence test.
Test candidates and keep a reproducible record
Before relying on a candidate, test it on simple systems relevant to the element and bonding you care about. The QE FAQ puts the advice plainly: “You should always test pseudopotentials on simple systems before trusting them!” Use the tests to examine whether the candidate behaves plausibly for your intended configurations, then establish convergence for the actual calculation.
When reporting results, record the precise UPF filename and version, source or library, XC functional, family, relativistic treatment, valence choices, cutoff settings, and convergence evidence. Preserve the file or a stable reference to it so the calculation can be reconstructed. QE asks users of externally generated pseudopotentials to credit their authors; include that attribution with the provenance details.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
The QE generation guide is useful for understanding how construction choices affect transferability, but its worked generator settings are examples of generation methodology, not current universal recommendations for files you should create or use.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
